On Fri, Mar 30, 2018 at 08:21:21AM -0700, Richard Cochran wrote:
> On Fri, Mar 30, 2018 at 05:02:14PM +0200, Andrew Lunn wrote:
> > Would it not be better to just call ethtool_get_ts_info() on the real
> > device? That would then also deal with the case that the 'real' device
> > is another virtual device stacked on top of a real device.
>
> That won't work.
>
> The returned 'ethtool_ts_info' is on the stack of
> ethtool_get_ts_info(). Both the top level and the inner call to
> will call copy_to_user(), but the top level will clobber the correct
> result with zeros.
Ah, O.K.
But it still seems like there should be one central place to handle a
device. Maybe factor out the part which deals with getting the
information from the device, from the part that copies it to
userspace?
Andrew
